The long and short of it

                                          Kazakhstani consumption at work, in Almaty.

How much do Kazakhstanis spend immediately, and how much over time?

The answers are in the table below. It uses annual data from the World Bank for Kazakhstan from 1990 through 2022. Kazakhstan became independent of the Soviet Union by 1992. 

Consumption and GDP (gross domestic product, the value of domestic production) are in 2015 international dollars, using purchasing power parity. In other words, these estimates adjust for inflation and for changes in exchange rates. An international dollar can buy the same product, in the local currency, anywhere around the world. Thus, if a hamburger cost $1 in the US in 2015 and 500 tenge in Kazakhstan today, the 2015 international dollar equals 500 tenge today. 

Time is the number of years that have passed since 1990, where Time = 1 for 1990. Since we control for the long-run time trend, we can interpret the coefficient of income as the amount of an additional dollar spent in the short run (that is, spent in less than a year). Of another dollar, the typical Kazakhstani quickly spent 95 cents as an average over 1990 through 2022. Had we analyzed 100 samples of the data, we would have found that the marginal propensity to consume varied from .843 to 1.05 across the samples. A marginal propensity to consume of over 1 may look impossible, but it just means that the Kazakhstani spent part of wealth as well as spending all income.  Note that “income” is money received this year, while “wealth” is money accumulated over the years.   

The Time coefficient indicates that spending dropped nearly 2 billion international dollars per year. Since Kazakhstan rose into the ranks of upper-middle economies over this period, and since the rich save a larger share of their income than the poor, this result is not surprising. To put it in perspective, note that GDP in 2022 was 222 billion international dollars. Thus Kazakhstanis save an average of roughly 1% of GDP per year, in addition to the 5 cents per dollar of income that they did not immediately spend. We interpret saving broadly: It includes the forced saving of tax payments as well as voluntary saving. 

Saving in Kazakhstan is a long-run phenomenon. This interpretation is reinforced by the negative intercept. This figure indicates that the typical Kazakhstani with zero income saved money, perhaps by letting interest accumulate on her wealth. 

If we do not control for the time trend, then the marginal propensity to consume drops from .946 to .645. In other words, the long-run marginal propensity to consume is lower than the short-run one. This is consistent with the interpretation of saving as long-run behavior. And this seems reasonable if we suppose that saving responds to information about financial markets, which takes time to understand, while consumption responds more to immediate needs and even to impulse. – Leon Taylor, Baltimore tayloralmaty@gmail.com 

Regression statistics 

R square .98 

Standard error 5.45 E^09 

Observations 33 

ANOVA       df     SS                  MS             F              Significance F 

Regression    2     4.31E+22      2.16E+22    725.056    0.000

Residual      30     8.93E+20      2.98E+19 

Total            32     4.4E+22 

Variable            Coeff.         T-stat         P-val         Lower 95%         Upper 95% 

Intercept          -7.9E+09     -3.06          0.005        -1.3E+10              -2.6E+09 

GDP                  0.946          18.63         0.000         0.843                    1.05 

Time                 -1.9E+09    -6.31          0.000        -2.5E+09              -1.3E+09

Notes 

For useful comments, I thank but do not implicate Annabel Benson and Mark Kennet. 

Reference 

World Bank. World Development Indicators. www.worldbank.org